The growing issue of adolescent myopia is mostly addressed through eye-focused methods, overlooking brain function's role. This study developed a toolbox using synchronized EEG and fNIRS to analyze brain activity ...
详细信息
ISBN:
(数字)9798331541750
ISBN:
(纸本)9798331541767
The growing issue of adolescent myopia is mostly addressed through eye-focused methods, overlooking brain function's role. This study developed a toolbox using synchronized EEG and fNIRS to analyze brain activity in adolescents during 2D and 3D visual training. The system integrates data collection, preprocessing, feature extraction, and visualization, capturing metrics like power spectral density, correlation, phase locking, and hemoglobin levels. It is easy to use, employs simple algorithms, and offers clear visuals, helping to understand the impact of visual training on brain function.
Modeling the moving behaviors and predicting the future paths of pedestrians, especially for those in complex scenes, remain a challenging problem in machine learning. We recognize that human motion trajectories, gove...
详细信息
Modeling the moving behaviors and predicting the future paths of pedestrians, especially for those in complex scenes, remain a challenging problem in machine learning. We recognize that human motion trajectories, governed by social norms and constrained by physical structures of the surrounding environment, are both forward predictable and backward predictable. Motivated by this observation, we develop a new approach, called reciprocal twin networks, for human trajectory learning and prediction. We design two networks, a forward prediction network to predict future trajectory from past observations and a backward prediction that performs the trajectory prediction backward in time. The backward prediction network serves as the inverse operation of the forward prediction network, forming a reciprocal constraint. During the training stage, this reciprocal constraint allows them to be jointly learned for accurate and robust human trajectory prediction. During the inference stage, we borrow the concept of adversarial attack of deep neural networks, which iteratively modifies the input of the network to match the given or forced network output, and develop a new method, called reciprocal attack for matched prediction, to achieve accurate human trajectory prediction. Our experimental results on benchmark datasets demonstrate that our new method outperforms the state-of-the-art methods for human trajectory prediction.
Scientific visualizations (SciVis) translate numerical and spatial data as images, enabling scientists to better understand the phenomena described by the data and gain insights that may be overlooked by statistical m...
详细信息
Teleoperation enables the control of robots in remote, complex, or hazardous environments such as manufacturing, healthcare, and disaster response. However, traditional teleoperation systems lack immersive depth perce...
详细信息
ISBN:
(数字)9798331514846
ISBN:
(纸本)9798331525637
Teleoperation enables the control of robots in remote, complex, or hazardous environments such as manufacturing, healthcare, and disaster response. However, traditional teleoperation systems lack immersive depth perception, which limits spatial awareness and precision. This paper presents a novel teleoperation framework integrating an Intel RealSense D345i depth camera with the Reachy humanoid robot, enhanced by augmented reality (AR) overlays in a virtual reality (VR) interface. By leveraging depth sensing, point cloud processing, and AR visualization, the system improves task accuracy, spatial awareness, and user experience. Our system utilizes advanced frameworks for seamless AR and VR integration. Preliminary findings demonstrate enhanced performance in tasks requiring precise object manipulation, with future user studies planned for quantitative evaluation.
The accurate measurement of the magnetizing inrush current in transformers is of the greatest interest for the correct operation and protection of power systems. This paper presents a method for analysis and visualiza...
详细信息
ISBN:
(纸本)9781728195391
The accurate measurement of the magnetizing inrush current in transformers is of the greatest interest for the correct operation and protection of power systems. This paper presents a method for analysis and visualization of time-varying harmonics using wavelets and its application to inrush currents in transformers. The Wavelet Packet Transform and the Inverse Wavelet Packet Transform are used as a filter bank for decomposition and reconstruction in the time domain of the harmonics in the inrush current. The peak and the r.m.s. magnitude and their decays, and the time evolution of harmonics in time and frequency domains as a function of the instant of connection of the input voltage to the primary winding are studied in detail. The results obtained can be used for the adequate design of the transformer protection and the selection of the instant of connection of voltage supply to the primary winding.
Advancements in technology have led to the virtual reality development in Indonesia, impacting organizations and businesses positively. Virtual reality offers opportunities for data visualization in the upcoming elect...
Advancements in technology have led to the virtual reality development in Indonesia, impacting organizations and businesses positively. Virtual reality offers opportunities for data visualization in the upcoming election, providing insights into candidate wins and vote distribution. This paper develops a virtual reality election data visualization application using Unity and Oculus Quest 2. It includes data acquisition, modelling, rendering, and user interactions. Usability testing was conducted with eight users using task-based scenarios.
In today's data-driven world, effective environmental management relies heavily on advanced analytics tools. This paper explores the application of Oracle Analytics Cloud for in-depth analysis and visualization of...
详细信息
ISBN:
(数字)9798350387681
ISBN:
(纸本)9798350387698
In today's data-driven world, effective environmental management relies heavily on advanced analytics tools. This paper explores the application of Oracle Analytics Cloud for in-depth analysis and visualization of environmental data to improve decision-making processes in the municipality of Kranj. The study aims to use Oracle Analytics Cloud for processing and analysing large data sets collected by the municipality, with a focus on detecting trends and patterns that could support policymaking. Oracle Analytics Cloud enables the integration of various data sources and provides a comprehensive platform for cross-functional environmental analysis.
Big data analysis is broader with a life cycle from choosing a model to represent data, generation of the model for the given data set, analysis using operations relevant to the model, and finally, visualization to un...
Big data analysis is broader with a life cycle from choosing a model to represent data, generation of the model for the given data set, analysis using operations relevant to the model, and finally, visualization to understand the results. The ability to perform this end-to-end workflow has become more important as the data size and complexity have increased significantly. visualization is especially important when the data sizes are large and models are complex (e.g., graphs, multilayer networks (MLNs)) which is difficult to comprehend without *** this poster paper, we describe a web-based tool MLN-Visualizer for visualizing graphs and MLN analysis results. Interactively, the user can visualize their uploaded input data and analysis results, in multiple ways, using a menu-driven interface. We hint at adding privacy to the resulting visualization.
Scene classification is a well-established area of computer vision research that aims to classify a scene image into pre-defined categories such as playground, beach and airport. Recent work has focused on increasing ...
详细信息
ISBN:
(纸本)9781728196817
Scene classification is a well-established area of computer vision research that aims to classify a scene image into pre-defined categories such as playground, beach and airport. Recent work has focused on increasing the variety of pre-defined categories for classification, but so far failed to consider two major challenges: changes in scene appearance due to lighting and open set classification (the ability to classify unknown scene data as not belonging to the trained classes). Our first contribution, SceneVLAD, fuses scene classification and visual place recognition CNNs for appearance invariant scene classification that outperforms state-of-the-art scene classification by a mean F1 score of up to 0.1. Our second contribution, OpenSceneVLAD, extends the first to an open set classification scenario using intra-class splitting to achieve a mean increase in F1 scores of up to 0.06 compared to using state-of-the-art openmax layer. We achieve these results on three scene class datasets extracted from large scale outdoor visual localisation datasets, one of which we collected ourselves.
In recent years, Deep Learning techniques have achieved some success in bioinformatics tasks, including protein conformation prediction. In this work, we propose a Bidirectional Long Short-Term Memory (BLSTM) network ...
详细信息
ISBN:
(纸本)9781665438278
In recent years, Deep Learning techniques have achieved some success in bioinformatics tasks, including protein conformation prediction. In this work, we propose a Bidirectional Long Short-Term Memory (BLSTM) network system, called Human Proteins Angles Prediction (HPAP), in order to improve the prediction of dihedral angles of proteins. We have introduced a discrete subdivision in classes of 5 degrees for protein torsion angles and four new features related to accessible surface area and volume. In total there are 73 classes (72 classes include the angles between -180 degrees and 180 degrees, a further class is used to code the free angles at the beginning of the sequence) with a maximum expected error of +/- 2.5 degrees. We have tested three model variants in several parameter combinations. With our model, we have obtained a decrease of the mean absolute error of about 2 degrees for the psi angle. Although our dataset is reduced in size, the accuracy of phi and psi angles is comparable to the existing methods. Predicting angles accurately is useful for accurately reconstructing the three-dimensional structure of a protein. In this context, the prediction is limited to the phi and psi angles and we will visualize what happens locally when a prediction is correct. In case the prediction is far from true angles, even a small error can deconstruct the backbone.
暂无评论